build the algorithms from scratch?
Every year, Kaggle runs a large-sample-size survey of Machine Learning & Data Science professionals. The 2020 results are out: https://www.kaggle.com/kaggle-survey-2020 … -- lots of great insights. Fun tidbit: a bit over half of ML & DS professionals use Keras.pic.twitter.com/7ppM4wbIkW
-
-
This Tweet is unavailable.
-
-
-
-
More power to you sir. Wish keras dominates the field of computer vision
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Pretty sure some picked Tensorflow but are mainly usint Keras :)
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Were Tensorflow/Keras linked in this survey, or did they both just happen to arrive at 50.5%? Looks quite surprising :)
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
lol, I misinterpreted your logo like pickachu holding news paper :)
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Hi Franscois, many times you mention that most practitioners prefer tf/keras, while many others make the opposite claim for pytorch. What's your take on that controversy? Are you just using different "metrics", e.g., Github stars, paper refs, kaggle stats, from one another?
-
That's a very valid and good question. These scores usually come as sums of unknown metrics. While the importance difference between these metrics can be helpful insight.
End of conversation
New conversation -
-
-
Why is it now sorted
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
I bet something like 25% of those scikit-learn imports are thanks to train_test_split
-
Couldn’t you use scikit to split the data but use TF or Keras to do the analytics?
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.